基于GEM-PHD粒子滤波的移动定位自更新传播算法  

A mobile localization self-updating propagation algorithm based on GEM-PHD particle filter

作　　者：黄庆东 姚雪茜 张淼 周赟 郝森 刘青 HUANG Qingdong;YAO Xueqian;ZHANG Miao;ZHOU Yun;HAO Sen;LIU Qing(School of Communications and Information Engineering,Xi'an University of Posts and Telecommunications,Xi'an 710121,China)

机构地区：[1]西安邮电大学通信与信息工程学院,陕西西安710121

出　　处：《西安邮电大学学报》2021年第1期13-19,共7页Journal of Xi’an University of Posts and Telecommunications

基　　金：陕西省重点创业创新链项目(2020ZDLGY02-06);国防科研试验信息安全实验室基础研究项目(2018XXAQ09)。

摘　　要：为了实现无线传感器网络(Wireless Sensor Networks,WSN)中移动节点的实时动态定位和更新,提出了一种基于广义运动概率假设密度粒子滤波的移动定位自更新传播算法,该算法利用锚节点作为观测者探测周围环境中存在的未知节点,收集探测到的信息,利用广义运动概率假设密度粒子滤波算法对未知节点定位。基于反向定位策略选择未知节点的最优位置,将已获得定位信息的节点升级为虚锚节点继续对周围邻居节点实施定位,再进行定位传播和位置更新,最终实现全网络移动节点实时定位和位置更新。仿真结果表明,在锚节点单一存在时,所提算法可以预估未知节点的位置,且定位精度高,可以对整个移动群体进行实时定位。In order to realize real-time dynamic positioning and updating of nodes in wireless sensor networks(WSN),a mobile positioning self-update propagation algorithm based on generalized motion probability hypothesis density particle filter is proposed.The algorithm first uses anchor nodes as observers to detect unknown nodes in the surrounding environment,which collects the detected information and uses the generalized motion probability hypothesis density particle filter algorithm to locate unknown nodes.Secondly,the reverse localization strategy is adopted to select the optimal location of the unknown nodes.Finally,upgrade those nodes that have already obtained location information as virtual anchor nodes to continue the localization process of the surrounding neighbor nodes,so as to achieve real-time localization propagation and location update of mobile nodes throughout the network.The simulation results show that the proposed algorithm estimates the location of the unknown node with higher accuracy and the whole mobile population can be positioned in real time when the anchor node exists alone.

关 键 词：无线传感器网络 移动节点定位 广义运动概率假设密度 数据融合 随机有限集 

分 类 号：TN929.5[电子电信—通信与信息系统]